Introduction

Coronary artery disease (CAD) which is a complex cardiovascular disease is the leading cause of death worldwide despite all measures [1]. Although risk factors such as smoking, diabetes, hypertension, triglyceride (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C) are identified, the changing prevalence of the disease at different ethnic groups pointing out the genetic background of CAD [2, 3]. In this context, the importance of identifying new candidate genes, and thus developing novel therapeutic approaches for atherosclerosis can be overemphasized. Since atherosclerosis is the major cause of CAD, lipid metabolism plays a central role in it, and genes related to cholesterol metabolism expected to be involved in the development of CAD [3].

Apolipoprotein D (ApoD) is mainly a component of HDL but also presents in very-low-density lipoproteins (VLDL) and LDL [4]. ApoD protein is widely expressed, unlike other apolipoproteins which are generally expressed in the liver [5]. Initial studies on ApoD showed that ApoD interacts with various ligands, including cholesterol [5]. Besides, previous reports were identified that the APOD gene polymorphisms are related to the characteristic lipid profile of metabolic syndrome [6]. The gene product of APOC1 is a constituent of chylomicrons, VLDL, and HDL and is involved in the regulation of the exchange of cholesterol between lipoproteins and removal of cholesterol from tissues [7, 8]. The low-density lipoprotein receptor (LDLR) is a cell surface protein that binds to the LDL which are the primary carriers of the cholesterol in the blood [9]. LDLR binds to the apoB particle on the surface of LDL and the receptor-ligand complex internalized by endocytosis [9]. Previous studies showed that the upregulation of LDLR protein decreases the risk of cardiovascular events by reducing the plasma LDL-C levels [10, 11]. Sortilin Related Receptor 1 (SORL1) which is also known as LDLR Relative with 11 Ligand-Binding Repeats (LR11) is a member of the LDLR family and overexpressed in atherosclerotic plaques [12]. In previous studies, soluble LR11 levels were found correlated with intima-media thickness in carotid arteries in dyslipidemic subjects and adverse clinical outcomes in CAD patients [13, 14]. Lysosomal acid lipase (LAL) which is encoded by the LIPA gene is the key lysosomal enzyme for hydrolyzing cholesteryl ester and triglycerides into free cholesterol [15]. In previous animal studies, when LDLR deficient mice are treated with targeted delivery of LAL to macrophages, the progression of the atherosclerotic lesion was prevented, and early lesions were eliminated [16].

In this study, it was aimed to determine the genetic association of coronary artery disease with cholesterol metabolism gene polymorphisms. For this aim APOC1 (rs11568822), APOD (rs1568565), LIPA (rs13500), SORL1(rs2282649), and LDLR (rs5930) gene polymorphisms were selected and the effects of these polymorphisms on CAD development and known metabolic risk factors such as plasma lipid levels and type 2 diabetes mellitus (T2DM) were investigated in Turkish CAD patients.

Material and methods

Study design and subjects

The study samples were recruited from Ufuk University Cardiology Department between 2014 and 2017; all samples were aged 18 years or over and underwent invasive coronary angiography due to stable angina pectoris, ischemia, and acute coronary syndrome. Selected 493 Turkish individuals [276 patients > 50% luminal stenosis narrowing and 217 non-CAD controls (< 30% stenosis)] were enrolled in the study (Supp. Table S1). Two independent operators were evaluated coronary luminal narrowing following the guidelines of the American College of Cardiology/American Heart Association (ACC/AHA) for the classification of coronary lesions. Patients with a history of previous coronary bypass graft surgery, ongoing decompensated (New York Heart Association Functional Class IV) heart failure, advanced hepatic or renal failure, a life expectancy of less than a year, and known malignancy (except basal cell carcinoma and squamous cell carcinoma of the skin in full remission), and those with an active infectious or inflammatory disease or a rheumatological disease were excluded from the study. Patients with a previous history of coronary brachytherapy and previous exposure to cholesteryl-ester transfer protein (CETP) inhibitors or PCSK9 inhibitors were also excluded from the study. SYNTAX (Synergy between percutaneous coronary intervention with Taxus and Cardiac Surgery) Scores and Gensini Scores were calculated for each sample following angiography to evaluate the complexity and severity of CAD. SYNTAX score which is used for the assessment of the complexity of CAD was calculated using an online calculator (version 2.28, www.syntaxscore.com) while the Gensini score was determined according to the degree of stenosis score, multiplying factor that depends on the region of the stenosis and collateral adjustment factor [17, 18]. For further analysis, the study group was classified according to the T2DM status with CAD status and according to sex independent from CAD status. The whole sample collection and analysis processes were conducted in compliance with the ethical guidelines of the Declaration of Helsinki and approved by the Institutional Review Board at Istanbul University (Approval no: 377879). The written informed consent was obtained from all participants and all experiments were performed following the approved guidelines and regulations.

Table 1 Genotype frequencies of selected polymorphisms stratified according to CAD, T2DM status, and sex
Full size table

Laboratory test

Blood serum samples were collected before coronary angiography and stored at − 80 °C until analyzed. Any lipemic, icterus, and hemolysis specimens were excluded. Analyses of biochemical parameters were performed. Concentrations of total cholesterol (TC), TG, fasting glucose, HDL-C, and LDL-C were measured by UniCel DxC 800 (Beckman Coulter, USA). Hemoglobin levels were analyzed using the High-Performance Liquid Chromatography (HPLC) system. The results were given as the ratio of glycosylated hemoglobin (HbA1c) to total hemoglobin in percentages.

Genetics analysis

Determination of the candidate genes genotypes

DNA was extracted from peripheral blood leucocytes using the inorganic method [19]. Selected 493 subjects (276 CAD and 217 non-CAD) were examined for APOC1 (rs11568822), APOD (rs1568565), LIPA (rs13500), SORL1(rs2282649), and LDLR (rs5930) polymorphisms. Genotyping was performed using hydrolysis probes in Real-Time PCR LightCycler® 480 (Roche Life Science, USA). DNA amplification was set up in 96-well plates (Roche Life Science, USA) with typical 10 μl PCR reaction consisted of 2 µl LightCycler® 480 Genotyping Master (Roche Life Science, USA) ready mix, 0.2 μl probes, and 0.2 μl primers, 6.6 μl distilled water. 1 μl (50 ng) genomic DNA was added to the PCR mixture. PCR was carried out using the following conditions: 95 °C for 10 min; 95 °C for 10 s, 56 °C for 30 s, 72 °C 1 s (45 cycles). The end-point analysis was performed for the discrimination of the alleles.

Statistical analysis

Hardy–Weinberg equilibrium was computed to the expected genotype distribution. The genotype distributions were compared using the chi-square test. Due to skewed distribution TG and HbA1c logarithmically transformed for analyses and expressed in geometric means. The two-tailed t-test was used to compare normally distributed continuous variables and expressed as means and standard deviation (SD), while categorical variables were compared using the chi-square test. Univariate analysis of variances adjusted for several factors was used for the determination of the association between polymorphisms and biochemical parameters and expressed in the mean and standard error of the mean (SEM). Logistic regression models were used to derive maximum odds ratio (OR) estimates and associated 95% confidence intervals (CIs), adjusted for several factors as confounders. A p-value < 0.05 was considered significant. All statistical analyses were performed using IBM SPSS (v23.0, IBM Corporation, USA).

Results

Baseline characteristics of study subjects

Baseline characteristics of CAD (mean age; 63.4 ± 10.6, 67.0% male) and non-CAD (mean age; 57.7 ± 11.7, 43.8% male) groups were compared, and results were summarized in Supp. Table 1. The frequencies of lipid-lowering and antidiabetic drug usages were higher in the CAD group compared to non-CAD (p < 0.01). Besides, the Gensini score, SYNTAX score, and percentage of stenosis were also higher in the CAD group when compared to non-CAD (p < 0.01, Supp. Table S1).

Associations of selected polymorphisms with T2DM

When the study population was grouped according to the T2DM status in addition to CAD and non-CAD classification, in the non-CAD group it was found that minor allele (C allele) carriage of APOD rs1568565 polymorphism is more prevalent in the non-T2DM group (p = 0.043, Table 1). For further analysis, the effect of APOD minor allele carriage status on T2DM was examined in CAD and non-CAD groups stratified according to sex. In the male non-CAD group, APOD minor allele carriage was observed in a higher percentage in non-T2DM (p = 0.022, Table 1). In a logistic regression analysis adjusted for BMI, age, and lipid-lowering drug usage, it was found that male minor allele carriers of APOD polymorphism had a lower risk for T2DM in the non-CAD group (OR = 0.302, [95% CI 0.111–0.821], p = 0.019, Table 2).

Table 2 The effects of the minor allele of the APOD and LDLR polymorphisms on T2DM stratified according to sex and CAD status
Full size table

Differences in LDLR rs5930 polymorphism genotype distributions (p = 0.050, Table 1) and allele frequencies (p = 0.017, Table 1) were observed among non-T2DM and T2DM groups in the male CAD group. Also, male CAD carriers of the minor allele of LDLR polymorphism were found in a higher percentage in the non-T2DM group (p = 0.016, Table 1). In logistic regression analysis adjusted for BMI, age, and lipid-lowering drug usage, it was found that minor allele carriage of LDLR polymorphism lowers the risk for T2DM (OR = 0.502 [95% CI 0.259–0.974] p = 0.042, Table 2).

There were no statistically significant differences between groups in genotype distributions and allele frequencies of selected APOC1, SORL1, and LIPA polymorphisms in CAD and non-CAD groups in analyses stratified according to T2DM status and sex (p > 0.05).

Associations between biochemical parameters and selected polymorphisms in CAD and T2DM subgroups

When the lipid profiles of individuals were examined, in a univariate analysis which is adjusted to age, sex, BMI, and lipid-lowering drug usage it was observed that APOD rs1568565 minor allele carriers have increased TC levels in non-CAD with T2DM group (p = 0.004, Fig. 1). In the univariate analysis with these same confounders, minor allele carriage of selected polymorphism of LIPA was found associated with elevated HDL-C levels in the non-CAD T2DM group (p = 0.031, Fig. 1).

Fig. 1
figure 1

Estimated mean and standard error of mean values for total cholesterol levels across the genotypes of APOD rs1568565 and HDL cholesterol levels across the genotypes of LIPA rs13500 polymorphisms were given in groups stratified according to CAD and T2DM status after adjustment for age, sex, BMI, and usage of lipid-lowering drugs. A Total cholesterol levels were found higher in C allele carrier T2DM patients of APOD rs1568565 polymorphism in the non-CAD group. B HDL cholesterol levels were elevated in LIPA rs13500 minor allele carrier T2DM patients in the non-CAD group

Full size image

Genotype distributions and allele frequencies of polymorphisms in CAD and non-CAD groups

Genotype distributions of APOC1 (rs11568822), APOD (rs1568565), LIPA (rs13500), SORL1 (rs2282649), and LDLR (rs5930) polymorphisms were given in Table 3. The genotype distributions and allele frequencies did not show statistically significant differences between non-CAD and CAD groups. The distributions of genotypes were in Hardy–Weinberg equilibrium for all polymorphisms.

Table 3 Genotype frequencies of selected polymorphisms
Full size table

Associations between biochemical parameters and selected polymorphisms in CAD and non-CAD

Association between genotype and lipid profiles which includes TC, LDL-C, HDL-C, and TG levels and other known risk factors such as BMI, HbA1c, and fasting glucose levels were examined in CAD and non-CAD groups. Minor allele carriage of APOC1 rs11568822 and LDLR rs5930 polymorphisms were found associated with decreased HbA1c and HDL-C levels in the non-CAD group, respectively (Supp. Table S2). In addition, an association was found between minor allele carriage of APOC1 polymorphism and elevated TG levels in the CAD group (p = 0.034, Supp. Table 2). When the study population classified according to sex independent from CAD status, male minor allele carriers of APOC1 polymorphism ha higher TG levels in a univariate analysis adjusted to age, BMI, and lipid-lowering drug usage (p = 0.003, Fig. 2). Besides, in a univariate analysis adjusted to age, BMI, lipid-lowering, and antidiabetic drug usage, the minor allele of LDLR polymorphism was found associated with higher HbA1c and glucose levels in females independent from CAD status (p < 0.05, Fig. 2). No statistically significant associations were observed between SORL1, LIPA, and APOD polymorphisms and biochemical parameters in CAD and non-CAD groups.

Fig. 2
figure 2

Estimated mean and 95% confidence intervals of A triglyceride levels across the genotypes of APOC1 rs11568822 in females and males, after adjustment for age, BMI, and usage of lipid-lowering drugs were given, triglyceride levels were found higher in rs11568822 minor allele carrier males. B Estimated mean and 95% confidence intervals of HbA1c and glucose levels in females and males were shown for LDLR rs5930 polymorphism. Both HbA1c and glucose levels were found higher in minor allele carrier females in the analysis adjusted to age, BMI, usage of lipid-lowering, and antidiabetic drugs

Full size image

Discussion

CAD is one of the most common cardiovascular diseases and is one of the leading causes of death at present [1]. Atherosclerosis is characterized by lipid accumulation, inflammatory response, cell death, and fibrosis in the arterial wall, which is the pathological basis for atherosclerotic CAD [3]. In this study, cholesterol metabolism-related gene polymorphisms were selected and studied among CAD patients and non-CAD controls. The results of the present study pointed out that APOD rs1568565, LDLR rs5930 polymorphisms are associated with T2DM status and its indicators such as HbA1c and glucose levels, while APOC1 rs11568822 and LIPA rs13500 polymorphisms are related to the lipid profile.

The associations between LDLR gene polymorphisms and CAD presence are widely studied among different ethnic populations [20,21,22]. On the other hand, there is a limited number of studies investigating the LDLR rs5930 polymorphism and they are mostly conducted in Alzheimer’s disease patients [23]. In this study, an association was found between rs5930 polymorphism and T2DM presence in a sex-specific manner. The rs5930 polymorphism has shown protective effects against T2DM in male CAD patients. Although there is no association between CAD, and T2DM presence and genotype distributions and allele frequencies in females, fasting glucose levels and HbA1c were found significantly higher in LDLR rs5930 minor allele carriers independent from CAD status. In a previous study that investigates the effect of another synonymous variant of the LDLR gene, rs688 polymorphism was found associated with cholesterol levels only in female individuals [24]. Zu et al. suggested that the rs688 polymorphism alters the exon splicing efficiency. Zu et al. pointed out that the association between cholesterol levels and the polymorphism is a result of alteration of the exon splicing efficancy [24]. According to RegulomeDB rs5930 polymorphism has a score of 0.61 while score 1 represents the variation is most likely regulatory [25]. It is possible that rs5930 polymorphism has a similar regulatory effect to rs688 in the region and could alter the activity of the gene. Besides, in previous reports, it was found that estradiol treatment positively regulates the expression of LDLR in vitro [26, 27]. Also, it was observed that postmenopausal women had lower levels of estradiol when compared to men [28]. The sex-specific result in the present study could be a result of the association between estradiol levels and LDLR expression. This study is the first study that shows the association between rs5930 polymorphism and its effects on T2DM status and disease-related parameters. The nature of the relation between LDLR and T2DM might be a result of the insulin receptor (IR) and LDLR complex. In previous studies, it was demonstrated that an intracellular co-association and co-localization exists for IR and LDLR in HepG2 and HUVEC cells [29, 30]. However, whether and how the rs5930 polymorphism affects the interaction of LDLR and IR remains unclear.

While the association between APOD rs1568565 polymorphism and CAD presence has not been previously investigated, the effects of other polymorphisms of APOD gene on CAD risk factors such as lipid profiles, T2DM, and insulin resistance were studied [6, 31, 32]. Also, animal studies indicated that the change of expression levels of APOD alters the insulin levels along with the lipid metabolism [33, 34]. In this study, the minor allele of APOD rs1568565 polymorphism was found associated with TC levels. In addition, the minor allele of rs1568565 polymorphism was found protective against T2DM. The mechanisms which lead to these results remain unclear and functional studies along with the studies with a larger sample size needs to be conducted for the confirmation of these associations.

The apoC-I protein is an important player in both lipid metabolism and inflammatory response. Many studies investigate the effects of gene polymorphisms in APOC1 on the development and progression of several different diseases including CAD, polycystic ovary syndrome, and Alzheimer’s disease [23, 35, 36]. Olsson et al. conducted a study in CAD patients and controls which indicates APOC1 rs11568822 polymorphism that also known as HpaI is associated with TG levels [36]. Also, in the same study, no association was found between genotype distributions of rs11568822 and CAD presence [36]. Similar to these results, in the present study minor allele carrier male subjects showed higher serum TG levels although no association was found between genotype distributions and CAD status. The reason for the higher TG levels could be the fact that rs11568822 polymorphism of APOC1 results in increased expression of the gene [37]. In a previous study conducted in the apoe−/− mice it has been demonstrated that APOC1 overexpression causes increased serum TG levels as a result of reduced LPL activity and increased VLDL production [38].

In this present study, no statistically significant differences in genotype distributions and allele frequencies of APOC1 rs11568822 and SORL1 rs2282649 polymorphisms were found between non-CAD and CAD groups. These results seem to be consistent with other research that found no association between APOC1 rs11568822 polymorphism and either CAD or myocardial infarction [36]. In accordance with the present results, another study that investigates the association between sortilin levels, inflammation markers, and SORL1 polymorphisms in coronary heart disease patients has demonstrated no difference in genotype distributions of rs2282649 polymorphism between groups [39]. In this study, no difference was found in allele frequencies and genotype distributions among CAD and non-CAD groups regarding APOD, LIPA, and LDLR polymorphisms. Although, this present study has a contribution to the literature and has importance since there were no previous studies to date that investigate the association between APOD rs1568565, LIPA rs13500, and LDLR rs5930 polymorphisms and CAD presence.

LAL deficiency as a result of the loss-of-function mutation in the LIPA gene was responsible for the lipid catabolism disorders such as Wolman disease and cholesteryl ester storage disease [40, 41]. In previous studies, an impaired ABCA1 protein expression along with the impaired cholesterol efflux was observed in LAL deficient individuals and these findings showed that LAL is one of the regulators of cholesterol efflux machinery and cholesterol homeostasis [42]. Also, in peritoneal macrophages isolated from lal−/− mice, cholesterol efflux was found in lower levels when compared to lal+/+ mice [16]. In the present study, we observed that minor allele carriers of LIPA rs13500 polymorphism have higher levels of HDL-C in the non-CAD T2DM group. This finding could be a result of the increased expression of the LIPA gene. The rs13500 polymorphism is located on the 3’ untranslated region of the gene and there are several microRNAs (miRNAs) that target this region according to prediction algorithms. The alteration of the miRNA binding site could affect the affinity of miRNA binding and therefore ultimately result in increased gene expression.

This present study is the first report that investigates the effects of the APOD rs1568565, LIPA rs13500, and LDLR rs5930 polymorphisms on CAD presence and disease risk factors. However, alongside the importance of the study, several limitations need to be considered. The effect of these five polymorphisms should be validated in a larger sample-sized study. In addition, there are no previous studies that investigate the functional effects of the polymorphisms and their implications on the expressions of the genes. Further research should be undertaken to investigate the possible mechanisms that underlie the findings of this study.

Conclusion

In conclusion, cholesterol-related gene polymorphisms influence lipid concentrations and the T2DM risk sex-specifically among Turkish CAD patients. We provided evidence that the LDLR rs5930 polymorphism is associated with T2DM presence only in the male CAD group. Besides, minor allele carrier females of LDLR polymorphism had elevated HbA1c and glucose levels independent from CAD status. In addition, the minor allele of APOD polymorphism was associated with decreased risk of T2DM in non-CAD. Also, APOD, LIPA, and APOC1 gene polymorphisms were demonstrated to affect serum lipid levels. Due to the importance of these processes in atherosclerotic CAD, we can conclude that cholesterol-related gene polymorphisms may play a crucial role during atherogenesis. Therefore, further research is sorely warranted in this direction.